W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > September 2001

Dealing with changing referents (was: Re: Working on glossary)

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Wed, 12 Sep 2001 17:11:35 -0500
Message-Id: <p05101007b7c589d6cb6e@[205.160.76.192]>
To: Graham Klyne <Graham.Klyne@MIMEsweeper.com>
Cc: w3c-rdfcore-wg@w3.org
OK, let me start with a clean whiteboard here.

The issue seems to be, that if URIs are 'names', ie refer to things, 
then what a given URI might refer to may change with time. And, 
moreover, that this isn't something to try to ban, but an intuitive 
and reasonable and useful thing. OK, lets accept that. Now, what do 
we do about it? Several strategies are possible.

1. Shelve it as an issue to be dealt with later, and ignore it for 
now. (Pat's favorite.)

2. Somehow incorporate it into the semantics of RDF. The immediate 
problem here is that RDF as it stands hasn't a shred of syntax which 
explicitly refers to times or changes or tenses or anything like 
that, so how do we reconcile the timeless nature of the language with 
the ever-labile nature of its referents?
   2a. By saying that the *real* referent doesn't change, but is 
something like the entire history of all the values it can have as 
time goes by. (GK's idea, right?)
   2b. By saying that the RDF MT refers to an 'instantaneous use' of 
RDF, ie to the meaning of RDF at some instant, and allow that this 
meaning might change when the same piece of RDF syntax is used a 
second time.
Moving between these seems to be like your 'currying' point, where 
the time-index gets shifted from the assertion (triplet) in case 2b 
to the URI (which then gets time-sliced) in case 2a.

3. Say that RDF is 'really' a modal logic, it just hasn't got any 
modalities defined yet, and so the MT corresponds to a 'temporally 
possible world'.  (This is one approach to how to do the 'later' in 
option 1, in fact).

4. Insist that RDF really is timeless and doesn't change, and so the 
URIs must be interpreted timelessly also, invoking some kind of 
integral dating scheme (Larry Masinter's idea).

-----

I can see some merit an all these ideas, but I guess that I still 
think that it is best to shelve this right now.  The issue isn't 
going to go away, but it seems to be outside the WG scope.

Pat

-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Wednesday, 12 September 2001 18:11:39 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:39:42 EDT